Social media platform X overhauls AI chatbot amid election accuracy concerns

In a letter to Elon Musk, they recommended redirecting users to vote.gov for accurate voting information. The five secretaries of state commended X’s update in a joint statement, acknowledging can i vote.org and  vote.gov as reliable sources for voters to connect with local election officials.

They expressed appreciation for X’s efforts to enhance their platform and encouraged continued improvements to ensure users receive accurate information from trusted sources, especially in this crucial election year, as reported by AP.

However, concerns remain regarding Grok’s capability to generate misleading AI-created images related to elections, which users have exploited to disseminate fabricated images of political figures like Kamala Harris and Donald Trump.

Grok, was launched last year for premium subscribers with Elon Musk’s promise that it would tackle “spicy questions” that other AI systems avoided.

However, Grok spread incorrect election information for 10 days before being corrected. Despite this, X remained silent on the issue.

Social media platforms are under growing pressure to address their role in spreading misinformation, especially during elections.

The secretaries of state’s letter pointed out that AI products like Grok, based on large language models, are prone to inaccuracies. Since Musk acquired Twitter (now X) in 2022, concerns have mounted about rising hate speech and misinformation.

Experts fear this trend undermines progress made in combating political disinformation since the 2016 US presidential election, potentially fueling misinformation ahead of the November elections.